| 1 | /* -*-c-*- |
| 2 | * |
| 3 | * ChaCha core definitions |
| 4 | * |
| 5 | * (c) 2015 Straylight/Edgeware |
| 6 | */ |
| 7 | |
| 8 | /*----- Licensing notice --------------------------------------------------* |
| 9 | * |
| 10 | * This file is part of Catacomb. |
| 11 | * |
| 12 | * Catacomb is free software; you can redistribute it and/or modify |
| 13 | * it under the terms of the GNU Library General Public License as |
| 14 | * published by the Free Software Foundation; either version 2 of the |
| 15 | * License, or (at your option) any later version. |
| 16 | * |
| 17 | * Catacomb is distributed in the hope that it will be useful, |
| 18 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 19 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 20 | * GNU Library General Public License for more details. |
| 21 | * |
| 22 | * You should have received a copy of the GNU Library General Public |
| 23 | * License along with Catacomb; if not, write to the Free |
| 24 | * Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, |
| 25 | * MA 02111-1307, USA. |
| 26 | */ |
| 27 | |
| 28 | #ifndef CATACOMB_CHACHA_CORE_H |
| 29 | #define CATACOMB_CHACHA_CORE_H |
| 30 | |
| 31 | #ifdef __cplusplus |
| 32 | extern "C" { |
| 33 | #endif |
| 34 | |
| 35 | /*----- Header files ------------------------------------------------------*/ |
| 36 | |
| 37 | #include <mLib/bits.h> |
| 38 | #include <mLib/macros.h> |
| 39 | |
| 40 | #ifndef CATACOMB_CHACHA_H |
| 41 | # include "chacha.h" |
| 42 | #endif |
| 43 | |
| 44 | #ifndef CATACOMB_SALSA20_CORE_H |
| 45 | # include "salsa20-core.h" |
| 46 | #endif |
| 47 | |
| 48 | /*----- Magic constants ---------------------------------------------------*/ |
| 49 | |
| 50 | /* The magic ChaCha constants are the same as the Salsa20 ones. For 256-bit |
| 51 | * keys ... |
| 52 | */ |
| 53 | #define CHACHA_A256 SALSA20_A256 /* e x p a */ |
| 54 | #define CHACHA_B256 SALSA20_B256 /* n d 3 */ |
| 55 | #define CHACHA_C256 SALSA20_C256 /* 2 - b y */ |
| 56 | #define CHACHA_D256 SALSA20_D256 /* t e k */ |
| 57 | |
| 58 | /* ... and for 128-bit keys ... */ |
| 59 | #define CHACHA_A128 SALSA20_A128 /* e x p a */ |
| 60 | #define CHACHA_B128 SALSA20_B128 /* n d 1 */ |
| 61 | #define CHACHA_C128 SALSA20_C128 /* 6 - b y */ |
| 62 | #define CHACHA_D128 SALSA20_D128 /* t e k */ |
| 63 | |
| 64 | /* ... and for 80-bit keys, for completeness's sake. */ |
| 65 | #define CHACHA_A80 SALSA20_A80 /* e x p a */ |
| 66 | #define CHACHA_B80 SALSA20_B80 /* n d 1 */ |
| 67 | #define CHACHA_C80 SALSA20_C80 /* 0 - b y */ |
| 68 | #define CHACHA_D80 SALSA20_D80 /* t e k */ |
| 69 | |
| 70 | /*----- The ChaCha core function ------------------------------------------*/ |
| 71 | |
| 72 | /* The ChaCha quarter round. Read from the matrix @y@ at indices @a@, @b@, |
| 73 | * @c@, and @d@; and write to the corresponding elements of @z@. |
| 74 | */ |
| 75 | #define CHACHA_QR(z, y, a, b, c, d) do { \ |
| 76 | (z)[a] = (y)[a] + (y)[b]; (z)[d] = ROL32((y)[d] ^ (z)[a], 16); \ |
| 77 | (z)[c] = (y)[c] + (z)[d]; (z)[b] = ROL32((y)[b] ^ (z)[c], 12); \ |
| 78 | (z)[a] = (z)[a] + (z)[b]; (z)[d] = ROL32((z)[d] ^ (z)[a], 8); \ |
| 79 | (z)[c] = (z)[c] + (z)[d]; (z)[b] = ROL32((z)[b] ^ (z)[c], 7); \ |
| 80 | } while (0) |
| 81 | |
| 82 | /* The ChaCha double-round. Read from matrix @y@, writing the result to |
| 83 | * @z@. |
| 84 | */ |
| 85 | #define CHACHA_DR(z, y) do { \ |
| 86 | CHACHA_QR(z, y, 0, 4, 8, 12); \ |
| 87 | CHACHA_QR(z, y, 1, 5, 9, 13); \ |
| 88 | CHACHA_QR(z, y, 2, 6, 10, 14); \ |
| 89 | CHACHA_QR(z, y, 3, 7, 11, 15); \ |
| 90 | CHACHA_QR(z, z, 0, 5, 10, 15); \ |
| 91 | CHACHA_QR(z, z, 1, 6, 11, 12); \ |
| 92 | CHACHA_QR(z, z, 2, 7, 8, 13); \ |
| 93 | CHACHA_QR(z, z, 3, 4, 9, 14); \ |
| 94 | } while (0) |
| 95 | |
| 96 | /* The ChaCha feedforward step, used at the end of the core function. Here, |
| 97 | * @y@ contains the original input matrix; @z@ contains the final one, and is |
| 98 | * updated. This is the same as Salsa20, only without the final permutation. |
| 99 | */ |
| 100 | #define CHACHA_FFWD(z, y) do { \ |
| 101 | int _i; \ |
| 102 | for (_i = 0; _i < 16; _i++) (z)[_i] += (y)[_i]; \ |
| 103 | } while (0) |
| 104 | |
| 105 | /* Various numbers of rounds, unrolled. Read from @y@, and write to @z@. */ |
| 106 | #define CHACHA_4R(z, y) \ |
| 107 | do { CHACHA_DR(z, y); CHACHA_DR(z, z); } while (0) |
| 108 | #define CHACHA_8R(z, y) \ |
| 109 | do { CHACHA_4R(z, y); CHACHA_4R(z, z); } while (0) |
| 110 | #define CHACHA_12R(z, y) \ |
| 111 | do { CHACHA_8R(z, y); CHACHA_4R(z, z); } while (0) |
| 112 | #define CHACHA_20R(z, y) \ |
| 113 | do { CHACHA_12R(z, y); CHACHA_8R(z, z); } while (0) |
| 114 | |
| 115 | /* Apply @n@ (must be even) rounds, rolled. (This seems to be faster, |
| 116 | * probably because it fits in cache better). Read from @y@, and write to |
| 117 | * @z@. |
| 118 | */ |
| 119 | #define CHACHA_nR(z, y, n) do { \ |
| 120 | int _i; \ |
| 121 | CHACHA_DR(z, y); \ |
| 122 | for (_i = 0; _i < (n)/2 - 1; _i++) CHACHA_DR(z, z); \ |
| 123 | } while (0) |
| 124 | |
| 125 | /* Step the counter in the Chacha state matrix @a@. */ |
| 126 | #define CHACHA_STEP(a) \ |
| 127 | do { (a)[12] = U32((a)[12] + 1); (a)[13] += !(a)[12]; } while (0) |
| 128 | |
| 129 | /*----- Variants and naming -----------------------------------------------*/ |
| 130 | |
| 131 | /* Common numbers of rounds, for which we generate definitions. */ |
| 132 | #define CHACHA_VARS(_) _(8) _(12) _(20) |
| 133 | |
| 134 | /*----- That's all, folks -------------------------------------------------*/ |
| 135 | |
| 136 | #ifdef __cplusplus |
| 137 | } |
| 138 | #endif |
| 139 | |
| 140 | #endif |